diff options
author | Jens Axboe <axboe@fb.com> | 2015-04-15 17:05:48 -0600 |
---|---|---|
committer | Al Viro <viro@zeniv.linux.org.uk> | 2015-04-24 15:45:28 -0400 |
commit | fe0f07d08ee35fb13d2cb048970072fe4f71ad14 (patch) | |
tree | beb614e8860cfa1791143d01ba17f686304c5caf /include/linux | |
parent | 8e3c500594dca9a12c27eb6d77b82e0766879bfd (diff) | |
download | linux-stable-fe0f07d08ee35fb13d2cb048970072fe4f71ad14.tar.gz linux-stable-fe0f07d08ee35fb13d2cb048970072fe4f71ad14.tar.bz2 linux-stable-fe0f07d08ee35fb13d2cb048970072fe4f71ad14.zip |
direct-io: only inc/dec inode->i_dio_count for file systems
do_blockdev_direct_IO() increments and decrements the inode
->i_dio_count for each IO operation. It does this to protect against
truncate of a file. Block devices don't need this sort of protection.
For a capable multiqueue setup, this atomic int is the only shared
state between applications accessing the device for O_DIRECT, and it
presents a scaling wall for that. In my testing, as much as 30% of
system time is spent incrementing and decrementing this value. A mixed
read/write workload improved from ~2.5M IOPS to ~9.6M IOPS, with
better latencies too. Before:
clat percentiles (usec):
| 1.00th=[ 33], 5.00th=[ 34], 10.00th=[ 34], 20.00th=[ 34],
| 30.00th=[ 34], 40.00th=[ 34], 50.00th=[ 35], 60.00th=[ 35],
| 70.00th=[ 35], 80.00th=[ 35], 90.00th=[ 37], 95.00th=[ 80],
| 99.00th=[ 98], 99.50th=[ 151], 99.90th=[ 155], 99.95th=[ 155],
| 99.99th=[ 165]
After:
clat percentiles (usec):
| 1.00th=[ 95], 5.00th=[ 108], 10.00th=[ 129], 20.00th=[ 149],
| 30.00th=[ 155], 40.00th=[ 161], 50.00th=[ 167], 60.00th=[ 171],
| 70.00th=[ 177], 80.00th=[ 185], 90.00th=[ 201], 95.00th=[ 270],
| 99.00th=[ 390], 99.50th=[ 398], 99.90th=[ 418], 99.95th=[ 422],
| 99.99th=[ 438]
In other setups, Robert Elliott reported seeing good performance
improvements:
https://lkml.org/lkml/2015/4/3/557
The more applications accessing the device, the worse it gets.
Add a new direct-io flags, DIO_SKIP_DIO_COUNT, which tells
do_blockdev_direct_IO() that it need not worry about incrementing
or decrementing the inode i_dio_count for this caller.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Elliott, Robert (Server Storage) <elliott@hp.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jens Axboe <axboe@fb.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Diffstat (limited to 'include/linux')
-rw-r--r-- | include/linux/fs.h | 29 |
1 files changed, 28 insertions, 1 deletions
diff --git a/include/linux/fs.h b/include/linux/fs.h index b1d7db28c13c..9055eefa92c7 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2635,6 +2635,9 @@ enum { /* filesystem can handle aio writes beyond i_size */ DIO_ASYNC_EXTEND = 0x04, + + /* inode/fs/bdev does not need truncate protection */ + DIO_SKIP_DIO_COUNT = 0x08, }; void dio_end_io(struct bio *bio, int error); @@ -2657,7 +2660,31 @@ static inline ssize_t blockdev_direct_IO(struct kiocb *iocb, #endif void inode_dio_wait(struct inode *inode); -void inode_dio_done(struct inode *inode); + +/* + * inode_dio_begin - signal start of a direct I/O requests + * @inode: inode the direct I/O happens on + * + * This is called once we've finished processing a direct I/O request, + * and is used to wake up callers waiting for direct I/O to be quiesced. + */ +static inline void inode_dio_begin(struct inode *inode) +{ + atomic_inc(&inode->i_dio_count); +} + +/* + * inode_dio_end - signal finish of a direct I/O requests + * @inode: inode the direct I/O happens on + * + * This is called once we've finished processing a direct I/O request, + * and is used to wake up callers waiting for direct I/O to be quiesced. + */ +static inline void inode_dio_end(struct inode *inode) +{ + if (atomic_dec_and_test(&inode->i_dio_count)) + wake_up_bit(&inode->i_state, __I_DIO_WAKEUP); +} extern void inode_set_flags(struct inode *inode, unsigned int flags, unsigned int mask); |